WellBeing International WBI Studies Repository

12-2008

Developmental Toxicity Testing: Protecting Future Generations?

Jarrod Bailey Physicians Committee for Responsible Medicine

Follow this and additional works at: https://www.wellbeingintlstudiesrepository.org/acwp_arte

Part of the Animal Experimentation and Research Commons, Animal Studies Commons, and the Other Anthropology Commons

Recommended Citation Bailey, J. (2008). Developmental toxicity testing: protecting future generations?. Alternatives to laboratory animals: ATLA, 36(6), 718-721.

This material is brought to you for free and open access by WellBeing International. It has been accepted for inclusion by an authorized administrator of the WBI Studies Repository. For more information, please contact [email protected]. 718 ATLA 36, 718–721, 2008

Developmental Toxicity Testing: Protecting Future Generations?

Jarrod Bailey

Physicians Committee for Responsible Medicine, Washington, DC, USA

Summary — A recent editorial is discussed, which implied that animal-based developmental and repro- ductive toxicology tests will continue to be crucial, that the thalidomide disaster could have been prevented by more animal testing, and that tests on juvenile animals would help to protect children (as developing adults) from the adverse effects of pharmaceuticals. It is argued that animal tests in these scientific areas do not provide reliable data that are predictive for human responses and, even if they did, the tests are too expensive and time-consuming for application to the very large number of substances that need to be tested. It is estimated there are already more than 100,000 man-made chemicals to which humans may be exposed on a regular basis, and it is therefore widely accepted that in vivo developmental toxicology could not possibly be used to assess all new and existing chemical substances, due to the scale of its demand upon time and resources. It is therefore imperative that alternatives such as those outlined above are embraced, further developed, accepted and used — as a matter of urgency.

Key words: animal testing, birth defects, developmental toxicity, teratology, toxicity testing.

Address for correspondence: J. Bailey, Senior Research Consultant, Physicians Committee for Responsible Medicine, 5100 Wisconsin Avenue, N.W., Suite 400, Washington, DC 20016, USA. E-mail: [email protected]

Introduction (grouped by drug class/chemical nature) revealed significant discordance: the positive predictability A recent Guest Editorial in Lab Animal Europe, ranged from 75% for the hamster, down to 40% for “Developmental and reproductive toxicity testing: a the rabbit, which also exhibited a false-negative potted history” (1) affirmed that “the ultimate goal rate of 40%. The mean positive predictability rate of developmental and reproductive toxicology test- in the six species most frequently used historically ing is to provide data for risk assessment… to pro- (mouse, rat, rabbit, hamster, primate, dog) was tect our future generations.” While this is not in less than 55%, and the number of equivocal results doubt, the article implied that this would involve a remained high at just under 25%. continued dependence on animal-based tests. Furthermore, there were 139 animal results However, this laudable goal cannot be achieved across different species for 35 individual sub- via such means. There is ample evidence to show stances positively linked with human teratogenic- that this ‘data for risk assessment,’ when obtained ity. Just over half (56%) of the animal results were from animal tests, is not suitable for this purpose positive. This poor predictability was underlined — it neither permits confident human-risk assess- by a US Food and Drug Administration (FDA) ment, nor serves to protect our future generations. report which detailed the responses of the mouse, rat, rabbit, hamster and monkey to 38 known human teratogens, in which the mean percentage Reliability and Concordance of of correct positives was only 60% (3). This report Animal Data also analysed 165 compounds known to be non-ter- atogenic in humans, for which the mean negative The data presented here summarise some of the predictive value for these five species was 54%. salient points of a comprehensive, systematic and non-selective study completed in 2005 (2), which examined the developmental toxicity results for Contemporary data — for two species, rat almost 1400 substances in 12 different species, by and rabbit using several databases and reference texts. For groups of known human teratogens, results from tests in the rat correlated with human classifications Historical data — across twelve species in 64% of cases; for the rabbit, this correlation was 40%. When this analysis was focused on 35 individ- An analysis of the responses of up to 12 animal ual substances known to be associated with human species to 11 groups of known human teratogens teratogenesis, the positive predictability for the rat Comments 719 was 61%, with 29% of the results falsely negative. — two ‘moderate-to-high risk’ human teratogens The rabbit was positively predictive in 41% of cases, produced positive results in both the rat and the but produced false-negative results for 56% of sub- rabbit. stances. A small number of equivocal results were obtained in both these species (Table 1, depicting our Based on these data, it can be argued that the tests analysis of results and classifications contained in in the rat and the rabbit are not sufficiently pre- Schardein’s book [4]). dictive to justify their use, and that the high rate The performances of the rat and rabbit in tera- of false-negatives also raises concern over their tology tests were further elucidated by examining human relevance and applicability. This lack of the results for the 20 chemicals used in the predictive power is also underlined by the statistic ECVAM validation studies on three non-animal that, of 3301 substances tested prior to 1993, 37% alternative methods for developmental toxicology were classified as definitely, probably or possibly (Table 2). Nine of these substances had a human teratogenic in animals, but fewer than 2.3% of risk classification with which to compare the rat these substances were linked to human birth and rabbit results: defects. (4).

— four ‘unlikely’ human teratogens produced two negative, one equivocal and one positive result Summary and Conclusions in the rat, and 3 negative and 1 positive result in the rabbit. The predictive nature of animal — three ‘minimal-to-small risk’ human teratogens developmental toxicity tests produced three positive results in the rat, and one positive in the rabbit (the other two were Contrary to Moxon’s assertions that animal-based not tested). developmental toxicology provides valuable data to

Table 1: The results from rat and rabbit teratology tests for known human teratogens

Total no. of True False substances + +/– – positive (%) negative (%)

Rat 31 19* 3 9 61% 29% Rabbit 27 11 1 15 41% 56%

The total numbers of results in the rat and rabbit for the 35 known human teratogens (categorised by Schardein [4]) are shown, followed by the number of positive (+), equivocal (+/–) and negative (–) conclu- sions. The final two columns reveal the percentages of rat and rabbit results that represented True Positives and False Negatives for these substances. * = one of these results was strain-dependent.

Table 2: The results from rat and rabbit teratology tests for nine chemicals used in ECVAM validation studies on non-animal alternative methods that had human risk classifications

Human teratogenic potential

Unlikely Minimal-to-small risk Moderate-to-high risk

++/–– ++/–– ++/––

Rat 1 1 2 3 2 Rabbit 1 3 1 2

The results for those nine substances for the rat and the rabbit, where they existed, are provided for chem- icals classified as posing ‘unlikely’, ‘minimal-to-small’ and ‘moderate-to-high’ human teratogenic risks. 720 Comments enable reliable human risk assessment, the exam- The use of lower organisms, embryo stages and ination of substantive data from decades of cell, tissue and organ cultures, was endorsed as animal-based teratology revealed significant vari- scientifically validated in 2001. The Embryonic ability in positive and negative predictability, and Stem Cell Test, for instance, uses two permanent high rates of false-positives, false-negatives and murine cell lines to screen for teratogenic potential equivocal outcomes across twelve species. These (8–12), and has scored highly on predictability, tests are therefore not suitable for their intended precision and accuracy in independent validation purpose. studies. It is already considered to be more repro- Further, while accepting the inherent scientific ducible, provide easier end-points, present no prob- and technical challenges involved, Moxon implies a lems with respect to ‘route of exposure,’ placental necessary role for juvenile animals in assessing the transfer and metabolic differences, and is devoid of safety of pharmaceuticals for children, based on an the confounding factors associated with animal acknowledgement of important differences tests, such as intra-species variability, environ- between children and adult humans and their sus- mental factors, differences in metabolism, placen- ceptibilities to the effects of pharmaceuticals. It tal and other anatomies, absorption, sensitivity, may be true that potential adverse events cannot metabolic activation, routes of administration, be assessed ethically and safely in paediatric clini- dose levels and strategies. It provides a means to cal trials, and that therefore some form of effective establish vital mechanistic models of teratogenic preclinical investigations must be conducted. action, via gene expression analysis, for example, However, one must be sceptical that animal tera- which will decrease the cost and increase the num- tology studies of any kind are fit for this purpose, ber of chemicals evaluated for developmental toxi- given the statistics presented here and their illus- city, could reduce the human impact of the tration of the lack of predictability of this false-positive and false-negative results generated approach. by animal models, and could also greatly reduce the numbers of animals used (2, 13). As human cell culture and other technologies improve, new proto- Alternative approaches to developmental cols will evolve, that will enable an even closer in toxicity tests vitro approximation of in vivo human teratogenesis (14). Fortunately, a number of alternatives to animal In addition, risk assessment can be aided by bet- testing exist or are in the course development, with ter information and data comparison and data the potential to improve the field of developmental sharing, and also by valuable human studies and toxicology in terms of time, cost and, most impor- birth-defect registries, which have identified many tantly, improved human predictability. Thus, it is important human teratogens (2, 15). anticipated that alternatives to the current ani- In conclusion, the article by Moxon (1), which mal-based methods will greatly enhance the num- implied that animal-based developmental and ber of substances that can be evaluated for reproductive toxicology tests will continue to be potential developmental toxicity, at lower cost and crucial, that the thalidomide disaster could have in a shorter time frame. It is estimated there are already more than 100,000 man-made chemicals to been prevented by more animal testing, and that which humans may be exposed on a regular basis tests on juvenile animals would help to protect (5), and it is therefore generally accepted that in children (as developing adults) from the adverse vivo developmental toxicology could not possibly be effects of pharmaceuticals, has little or no scien- used to assess all the new and existing chemical tific basis. Animal tests in these scientific areas do substances due to the scale of its demands upon not provide reliable data that are predictive for time and resources (6). human responses and, even if they did, the tests Computer-based systems, such as expert sys- are too expensive and time-consuming for applica- tems and structure–activity relationship (SAR) tion to the very large number of substances that analyses, and physiologically based pharmacoki- need to be tested. It is therefore imperative that netic modelling (PBPK), have already been respon- alternatives such as those outlined above are sible for the elimination of many animal tests in embraced, further developed, accepted and used — the pre-screening of candidate drug compounds. as a matter of urgency. The recently established US Environmental Pro- tection Agency (EPA) ToxCast™ programme employs the computational modelling of high- Acknowledgement throughput screening data to assist in prediction of the potential toxicities of chemicals to humans, This essay, and the original study on which it is incorporating reproductive endpoints, as well as based, was funded by the Physicians Committee other endpoints, including a variety of biochemical for Responsible Medicine (PCRM), Washington and developmental assays (7). DC, USA. Comments 721

References F., Bremer, S., Becker, K. & Spielmann, H. (2000). Development of prediction models for three in vitro 1. Moxon, M. (2008). Developmental and reproductive embryotoxicity tests in an ECVAM validation toxicity testing: a potted history. Lab Animal study. In Vitro Molecular Toxicology 13, 51–66. Europe 8(3), 3. 9. Genschow, E., Scholz, G., Brown, N.A., Piersma, 2. Bailey, J., Knight, A. & Balcombe, J. (2005). The A.H., Brady, M., Clemann, N., Huuskonen, H., future of teratology research is in vitro. Biogenic Paillard, F., Bremer, S. & Spielmann, H. (1999). Amines — Stress & Neuroprotection 19,97–145. Development of prediction models for three in vitro Avail able at: http://www.curedisease.net/reports/ embryotoxicity tests which were evaluated in an teratology.pdf (Accessed 14.07.08). ECVAM validation study. ALTEX 16, 73–83. 3. Anon. (1980). United States Food and Drug Admin- 10. Scholz, G., Pohl, I., Genschow, E., Klemm, M. & Spiel- istration. Caffeine: deletion of GRAS status, pro- mann, H. (1999). Embryotoxicity screening using posed declaration that no prior sanction exists, and embryonic stem cells in vitro: correlation to in vivo use on an intern basis pending additional study. teratogenicity. Cells, Tissues, Organs 165, 203–211. Federal Register 45, 69,817. 11. Scholz, G., Genschow, E., Pohl, I., Bremer, S., 4. Schardein, J.L. (1993). Chemically Induced Birth Paparella, M., Raabe, H., Southee, J. & Spielmann, Defects, 902pp. New York, NY, USA: Marcel H. (1999). Prevalidation of the Embryonic Stem Cell Dekker. Test (EST) — A new in vitro embryotoxicity test. 5. Anon. (2006). European Parliament Memo. Q & A Toxicology in Vitro 13, 675–681. on the new chemicals policy, REACH. MEMO/ 12. Spielmann, H., Pohl, I., Döring, B., Liebsch, M. & 06/488, 16pp. Brussels, Belgium: European Moldenhauer, F. (1997). The embryonic stem cell Parliament. Available at: http://europa.eu/rapid/ test (EST), an in vitro embryotoxicity test using two pressReleasesAction.do?reference=MEMO/06/488 permanent mouse cell lines: 3T3 fibroblasts and (Accessed 12.11.08). embryonic stem cells. In Vitro Toxicology 10, 6. Kotwani, A., Mehta, V.L. & Iyengar, B. (1995). 119–127. Aspirin by virtue of its acidic property may act as 13. Adler, S., Pellizzer, C., Hareng, L., Hartung, T. & teratogen in early chick embryo. Indian Journal of Bremer, S. (2008). First steps in establishing a Physiology & Pharmacology 39, 131–134. developmental toxicity test method based on 7. Anon. (2008). ToxCast™ Program: Predicting human embryonic stem cells. Toxicology in Vitro 22, Hazard, Characterizing Toxicity Pathways, and 200–211. Prioritizing the Toxicity Testing of Environmental 14. Adler, S., Lindqvist, J., Uddenberg, K., Hyllner, J. Chemicals. Research Triangle Park, NC, USA: & Strehl, R. (2008). Testing potential developmen- National Centre for Computational Toxicology, tal toxicants with a cytotoxicity assay based on EPAS. Available at: http://www.epa.gov/comptox/ human embryonic stem cells. ATLA 36, 129–140. toxcast/ index. html (Accessed 08.10.08). 15. Polifka, J.E. & Friedman, J.M. (1999). Clinical ter- 8. Genschow, E., Scholz, G., Brown, N., Piersma, A., atology: identifying teratogenic risks in humans. Brady, M., Clemann, N., Huuskonen, H., Paillard, Clinical Genetics 56, 409–420.